Claim-Level Evidence for Physics-Informed Machine Learning in Resilient Microgrids

PI-MG-Transitions Claim-Level Evidence for Physics-Informed Machine Learning in Resilient Microgrids This repository contains the complete coded evidence corpus used in the paper: Physics-Informed Machine Learning for Resilient Microgrid Dynamics and Control The repository enables full replication, audit, and reuse of the claim-level synthesis presented in the paper. What this repository contains Each claim corresponds to one method evaluated over one explicitly defined transition window, contingency regime, and enforcement locus. Claims, not papers, are the unit of evidence. The dataset includes: Mapping from bibliographic works to claim units Transition definitions (islanding, emergency islanded operation, restoration, reconnection) Physics-injection mechanism classifications Control-layer task assignments Declared operational limits and stability semantics Quantitative constraint and recovery outcomes Robustness stress tests (topology, disturbance, observability, communication) Validation fidelity tiers (F0–F3) Auditability tiers (A0–A3) Inter-rater annotations and reconciliation records